proof

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXFILTRATION]: The skill transmits document content to the official domain proofeditor.ai and interacts with a local service on localhost:9847. These network operations are documented and essential for the skill's primary function as a collaborative document editor.
  • [COMMAND_EXECUTION]: Uses Bash to perform API calls with curl and process JSON data with jq. These operations are used safely for the intended purpose of document management and do not involve arbitrary command execution.
  • [PROMPT_INJECTION]: The skill processes markdown documents from external sources, representing a potential surface for indirect prompt injection.
  • Ingestion points: Fetches markdown data from proofeditor.ai API endpoints and the local bridge at localhost:9847.
  • Boundary markers: No explicit delimiters or instructions are used to distinguish document content from agent instructions.
  • Capability inventory: The skill utilizes Bash (for curl/jq), Read, Write, and WebFetch capabilities.
  • Sanitization: There is no mention of sanitizing or validating the retrieved markdown content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 09:55 AM